Our long-term goal is to prevent death and disability from thrombotic events. Recently, we have been studying how to manage and dose Warfarin therapy and how to prevent stroke in patients who have atrial fibrillation. A major problem with prescribing Warfarin therapy is that its narrow therapeutic index and "idiosyncratic" response conspire to make its initiation protracted and dangerous. The proposed study focuses on how genetic and clinical factors can be used to predict this response. The primary objective of this 2-year proposal is to construct, validate, and use a warfarin-dosing algorithm for initiation of therapy. The secondary objective is to validate the paradigm that knowledge of pharmacogenetics can decrease the toxicity of commonly prescribed drugs. The study has 4 aims: 1. To construct a genetic-based, warfarin-dosing algorithm that will guide warfarin induction. To develop the algorithm, we will use clinical data and genetic specimens that we have obtained from 298 patients taking chronic warfarin therapy. We propose to complete our genotyping of these specimens for single nucieotide polymorphisms (SNPs) of the cytochrome P450 2C9 gene. We will test Hypothesis 1: The warfarin dosing algorithm will explain at least 30% of the variance in the therapeutic dose. 2. To validate the genetic-based, warfarin-dosing algorithm in an independent sample of 250 patients. We will test Hypothesis 2: The dosing algorithm will explain at least 30% of the variance in the therapeutic warfarin dose in the validation sample. 3. To use genetic-based warfarin dosing in 50 patients beginning warfarin therapy. We will use the genetic- based algorithm to initiate warfarin therapy in a pilot study of 50 patients. We will test Hypothesis 3: As compared to historic controls whose warfarin is initiated empirically, participants who use genetic-based warfarin dosing will be prescribed their therapeutic dose sooner. 4. To correlate genetic and clinical factors with the plasma level of free warfarin. Using high performance liquid chromatography, we will measure the levels of the 2 enantiomers of warfarin and test Hypothesis 4: Clearance of S-warfarin will be significantly decreased in the presence of 2C9 SNPs.